app.py

import gradio as gr
from youtube_transcript_api import YouTubeTranscriptApi
import re

def get_video_id(url):
    """Extract video ID from YouTube URL"""
    try:
        if "youtu.be" in url:
            return url.split("/")[-1]
        else:
            return re.findall(r"v=([^&]+)", url)[0]
    except Exception:
        raise ValueError("Invalid YouTube URL")

def get_transcript(url):
    """Get transcript for a YouTube video"""
    try:
        # Extract video ID from URL
        video_id = get_video_id(url)
    
        
       
        transcript = YouTubeTranscriptApi.get_transcript(video_id)
        # Get transcript
        
        
        # Combine all transcript pieces into one text
        full_transcript = "\n".join([item["text"] for item in transcript])
        
        return full_transcript
    
    except ValueError as e:
        return f"Error: {str(e)}"
    except Exception as e:
        return "Error: This video doesn't have captions available. Please try a different video."

# Create the Gradio interface
demo = gr.Interface(
    fn=get_transcript,
    inputs=gr.Textbox(
        label="YouTube URL",
        placeholder="Enter YouTube video URL here...",
    ),
    outputs=gr.Textbox(
        label="Transcript",
        lines=20,
        placeholder="Transcript will appear here..."
    ),
    title="YouTube Transcript Extractor",
    description="Enter a YouTube URL to get its transcript/captions.",
    examples=[
        ["https://www.youtube.com/watch?v=dQw4w9WgXcQ"],
        ["https://youtu.be/dQw4w9WgXcQ"]
    ]
)

if __name__ == "__main__":
    demo.launch()